Research Scientist, Video Foundation Models

Research Scientist focused on developing next-generation native video and omni foundation models for multimodal generation and understanding. This role involves large-scale model development across the full lifecycle, including architecture, data, training, evaluation, and efficient inference for high-quality, controllable video generation.

Skills and focus areas

• foundation models
• video generation
• multimodal generation
• multimodal understanding
• large-scale model development
• pre-training
• continued training
• post-training
• model architecture
• training objectives
• conditioning mechanisms
• data curation